Fractures involving both the posterior column and posterior wall of the acetabulum are among the most common pelvic fractures. The involvement of the articular surface and the mobility of the posterior column fragment make optimal reduction and fixation hard to achieve¹⁾.

Fixation techniques include fixation using screws or using plates and screw. The proximity of the joint surface to the posterior wall fragments makes fixation difficult due to the risk of joint penetration or impingement of the fractured fragment upon insertion of the screw. These difficulties lead to the use of spring plates. A spring plate is attached to the posterior column with one or two cortical screws and can be secured by buttressing by means of a reconstructed plate. The use of a spring plate avoids the risk of joint penetration and further impingement of the fragile fracture fragments²⁾.

Until now, there has been no acetabular plate design that would combine the spring plate’s function and the reconstruction plate’s function into one. A plate-like design would reduce operating time and costs. In this study, we aimed to assess the design of a new implant model that integrates the functionality of both the spring and reconstruction plates into a single unit using computerized finite element analysis (FEA). We chose this combination because it can support the posterior column and posterior wall of the acetabulum. We expect that this design will facilitate the treatment of combined posterior column fractures and posterior acetabular wall fractures, ease the fixation of the combined posterior column and posterior wall fracture and help reduce surgery time³⁾. Researchers aim to assess the design of a novel acetabular plate that integrates the functionality of both the spring plate and the reconstruction plate into a single unit using FEA. This design is expected to ease the fixation process of the combined posterior column and posterior wall fracture and help reduce surgery time.

This research conducted in Institut Teknologi Sepuluh Nopember in Indonesia. This was an observational study that compares seven pelvic models: normal pelvis, pelvis with posterior wall acetabular fracture, posterior column fracture, a combination of the above fractures, and the three fracture models fixed with the new acetabular plate. The new acetabular plate was designed as an anatomical plate for posterior wall and posterior collum fractures. Its design is similar to that of a posterior wall acetabular plate but with addition extension on the superior and inferior sides of the plate. This is useful for increase plate stability and providing an alternative to a bone purchase. This study aimed to compare the outcome of each condition and not to simulate the real-life surgery; therefore, we adjusted some of the most important assumptions (as described on the following paragraph) for this computer model to work.

Abaqus software (Simulia) was used to create a three-dimensional (3D) model of the pelvis. The pelvic bone model used 3D data from normal pelvic bone computed tomography scans. Ansys R23.2 program (Ansys Inc.) was used for FEA analysis. Posterior wall fracture occurs at about 90° of the posterior side of the acetabulum with complete detachment from the acetabulum, which is held only by soft tissues. Posterior collum fracture begins at the apex of the greater sciatic notch and goes through the articular and quadrilateral surfaces.

The bone model was assumed to be homogenous and isotropic, and to have linear elasticity. The whole bone was assumed to be cancellous with a Young’s modulus of 150 MPa and a density of 1.30 kg/mm³. A friction constant of 0.46 was used⁴⁾. The plates and screw material properties were based on those of titanium alloy, with a Young’s modulus of 107 MPa, density of 4.41 kg/mm³, yield strength of 1.098 MPa, and ultimate tensile strength of 1.33222 MPa. The plate model is depicted in Fig. 1⁴⁾. The plate was fitted to the contour of the pelvis. This plate will be fixated using three screws were on the superior side of the plate and two screws on the inferior side.

Figure 1. Implant design and its intended application in the pelvis.

In convergence analysis, we found stable deformation of elements larger than 0.08 mm (Fig. 2). We used 0.1 mm as our element size. The mean±standard deviation skewness was 0.28±0.18.

Figure 2. (A) Convergence analysis graph. (B) Visualization of discretized model of the pelvis.

The FEA analysis was performed by analyzing stress and deformation distribution in each pelvic model under a force of 1,000 N directed at 45° from the sagittal and coronal planes (Fig. 3). This force was chosen to simulate the force acting on the acetabulum in a standing position, as described previously⁵⁾. The pelvis was fixed on the sacroiliac joint and pubic symphysis⁵⁾.

Figure 3. Force direction on the pelvis.

In the normal pelvis, the largest deformity was found in the ischial tuberosity (3.91 mm) and the second largest on the inferior side of the acetabulum with a deformity (up to 3 mm). Stress distribution tended to be homogenous throughout the pelvis (Fig. 4).

Figure 4. Finite element analysis simulation of the normal pelvis (A, B) and the pelvis with a posterior column fracture (C, D). (A, C) Deformation evaluation. (B, D) Stress distribution.

In the posterior column fracture model (Fig. 4), the deformation was located mostly in the fractured acetabular part and gradually diminished in the ischiopubic ramus. The greatest deformation was 24.8 mm. In the other parts of the bone, the deformation was minimal. Given that the fixation or the immobile part of the pelvis is in the pubic and superior posterior iliac spine, while the force was directed into the acetabulum, deformation was certain to occur in the most mobile part and most stress should have been on the pivot point of the deformation. In this case, the most mobile part was the smaller fragment of the acetabulum and the pivot point was the inferior ramus.

In the posterior wall fracture model (Fig. 5), stress and deformation affected mostly the fractured fragment. Due to the software limitation, the acetabular fragment must remain connected to the pelvis and cannot completely detach from it. The most mobile part in this case was the fractured posterior wall and therefore the deformation would certainly happen in the detached part with a limited concentration of stress because there was no pivot point. In the combined fracture model, stress was greatest on the fracture site and was evenly distributed from the fracture site to the ischiopubic ramus, similar to the posterior wall fracture (Fig. 6).

Figure 5. Finite element analysis simulation of the pelvises with a posterior wall fracture (A, B) and with a combination of posterior wall and posterior column fractures (C, D). (A, C) Deformation evaluation. (B, D) Stress distribution.

Figure 6. Finite element analysis simulation of the pelvises with a posterior column fracture (A, B) and with a posterior wall fracture (C, D), both fixed with the new acetabular plate. (A, C) Deformation evaluation. (B, D) Stress distribution.

Upon fixation using the new acetabular plate, the deformation distribution in the fractured pelvis was normalized to resemble that in a normal pelvis. The largest deformation was in the middle of the plate and in the screw involved. Stress distribution was also normalized to resemble that in a normal pelvis, with the highest stress on the ischial tuberosity and the inferior side of the acetabulum. The largest deformation (approximately 0.8 mm) was similar in the three models and was observed in the middle of the plate (Fig. 6, 7).

Figure 7. Finite element analysis simulation of the pelvis with posterior wall and posterior column fractures fixed with the new acetabular plate. (A) Deformation evaluation. (B) Stress distribution.

Unlike simulating stress distribution in vivo, FEA can simulate the biomechanical environment of the pelvic ring and predict weight transfer and stress distribution without the involvement of various confounding factors. Our study elicited the stress distribution of the seven models and proved the efficacy of the new plate in fixating the fracture.

This study is in line with research conducted by Li et al.⁶⁾, who stated that the greatest stress occurs at the attachment point of the posterior acetabulum wall to the pelvis. This is reasonable, considering that this area is also the hinge/turn point of the posterior wall fragment from the acetabulum to the pelvic bone. On the basis of the general biomechanic of the hip, according Li et al.⁶⁾, the blue-green area is distributed around the sacroiliac joint and ischiopubic ramus in a lying posture. However, the area of stress distribution is different from our studies; it is clear that larger degrees indicate a larger area of pressure around the sacroiliac joint. According to Li et al.⁶⁾, finite element simulations can better reveal the biomechanical environment of the hip ring and predict load transfer and stress distribution between the fixation and bone structure. According to Li et al.⁶⁾, the pelvis is more stable in a vertical posture, meaning that sitting up in bed may be better than lying down for fracture healing.

The spring plate must be applied carefully to ensure that its hook can hold the acetabular rim without puncturing the labrum and does not scratch the head of the femur. The edge of the spring plate should be on the edge of the acetabular lip. The correct placement of the hook can be ensured with the help of a C-arm. Unless absolutely necessary, we recommend preserving the labrum and not opening the joint capsule. Therefore, the spring plate was positioned under the definitive posterior wall plate. The posterior wall of the acetabulum should be fixed with a lag screw, which is usually easy to insert. The glide hole can be drilled before reduction to ensure proper placement. One lag screw fixation performed on the posterior column acetabulum is to support the 3.5-mm reconstruction plate from the ischial tuberosity to the superior aspect of the acetabulum, taking care to keep the screw away from the joint.

In this study, simulations were also carried out with the same magnitude and direction of force but on fractures fixed with the new plate. In terms of deformation, the installation of this plate succeeded in preventing the posterior shift of the acetabulum wall. The minimal deformation that occurs is 0.37 mm. In terms of stress, almost all stress was redistributed to the acetabular plate, with almost none on the pelvic bones; more precisely, the highest stress was in the plate screw.

Deng et al.⁷⁾ also analyzed a new plate model in the pelvis but in the case of the anterior column–posterior hemitransverse fracture (ACPHTF). Their plate model, called the novel anatomical locking guide plate (NALGP), has screws for the anterior column and magic screws for fractures of the anterior column and acetabular hemitransverse posterior. Deng et al.⁷⁾ reported two main results: (1) compared to the other two groups, NALGPs had greater stiffness, especially when they were under loads higher than 600 N; (2) compared to the locking plate with posterior column screws, both the new plate and the double-locking plate provided better stability in fixing the ACPHTF. This novel plate was on par with the double-locking plate, especially regarding maximum overall displacement. Deng et al.⁷⁾ compared the stress distribution between the NALGP and screws and conventional plates and screws. Stress distribution can be interpreted as the ability of the plate or screw to withstand elastic deformation when subjected to a force because stress concentration at a certain point tends to cause plate or screw deformation and even damage. Whether for the force was 200 N, 400 N, or 600 N, the nephogram diagram revealed that the plate and screw stress distribution with NALGP was more uniform than with the other two plate models; however, there were areas of higher stress on the plates and screws even with NALGP⁷⁾.

Several mechanical studies^5-7) have compared different methods of fixing posterior wall fractures. One such study, conducted on cadaveric specimens, examined transverse and comminuted fractures. The addition of spring plates to the reconstruction increased the construct’s stiffness and strength. Another cadaveric biomechanical study comparing fixation methods found no significant differences in displacement among three methods: two lag screws only, two screws with a reconstruction plate, and two screws with a locked reconstruction plate.

FEA is a versatile tool widely used in engineering for predicting force and stress pressure, but it comes with several limitations. FEA operates under the assumptions that might not accurately represent nonlinear materials. Correctly defining boundary conditions is crucial, as unrealistic ones can lead to erroneous outcomes. Selecting appropriate element types is also vital for accuracy, as using unsuitable elements can introduce errors. Convergence issues may arise, particularly in nonlinear or complex problems, due to factors such as mesh distortion or numerical instability.

In conclusion, the use of FEA in biomechanical studies offers invaluable insights into the stress distribution and load transfer within pelvic structures during fracture fixation. From this study, it is obvious that the newly proposed plate is able to reduce, fix, and return stress distribution back to its normal state. Further in vivo and in vitro studies are warranted given this promising FEA result.

No funding to declare.

The authors thank the staff at the Department of Orthopedics and Traumatology of Dr. Soetomo General Hospital for their support and input during the writing of this manuscript.

No potential conflict of interest relevant to this article was reported.